Disentangling the influence of hydroclimatic patterns and agricultural management on river nitrate dynamics from sub-hourly to decadal time scales
Graphical abstract
Introduction
Production of synthetic nitrogen (N) fertilisers through the Haber-Bosch process and their application to arable fields have significantly contributed to increased food production worldwide (Galloway et al., 2004). However, they have also increased the amount of reactive N in the environment, which can subsequently have adverse effects on terrestrial and aquatic ecosystems, the Earth's climate and human health (Galloway et al., 2003, Sutton et al., 2011, Vitousek et al., 1997). Most N emissions in water in Western countries come from diffuse agricultural sources (Bouraoui and Grizzetti, 2011, Dupas et al., 2013) and nitrate is the dominant form transferred in intensively farmed areas (Mellander et al., 2012, Molenat et al., 2008, Rodriguez-Blanco et al., 2015).
National, federal or international regulations have been set up worldwide to reduce nitrate-N transfer to water. In the European Union, the Nitrate Directive (CEC 1991) and the Water Framework Directive (CEC 2000) provide guidelines for member states for national programmes of measures and evaluation protocols. Evaluation of long-term changes in nutrient concentrations in European rivers has highlighted little improvement in nitrate-N concentrations, despite implementation of the Nitrate Directive (Bouraoui and Grizzetti, 2011). One reason for this limited response to mitigation measures is the time lag due to the multi-year transfer time of nitrate-N in the unsaturated and saturated zones of catchments (Fovet et al., 2015, Howden et al., 2010, Melland et al., 2012, Windolf et al., 2012, Worrall et al., 2009, Wriedt and Rode, 2006). To assess the effectiveness of mitigation measures, long-term water quality records are thus necessary (Bouraoui and Grizzetti, 2011), firstly because of this time lag, and secondly because catchments are influenced by large-scale weather and climate patterns which can affect river nitrate-N concentrations (Gascuel-Odoux et al., 2010). If not evaluated properly, the effect of adaptive agricultural management may be confounded by the effect of multi-annual weather patterns. In addition, the dynamics of stream chemistry are typically non-linear and non-self-averaging over existing observation time lengths (Aubert et al., 2014, Kirchner and Neal, 2013); thus, non-stationary techniques such as Dynamic Harmonic Regression (DHR) are necessary to explore the data (Halliday et al., 2012, Halliday et al., 2013, Minaudo et al., 2015).
In recent years, interpretation of long-term water quality time series has provided useful insight into the water-quality trajectory of major river catchments, such as the Thames (Howden et al., 2010), the Seine (Billen et al., 2007, Romero et al., 2016), the Loire (Minaudo et al., 2015), the Elbe (Lehmann and Rode, 2001) and Baltic rivers (Andersen et al., 2015). Meanwhile, the development of high-frequency sensors has upgraded monitoring resolution to the time scales of fundamental processes, i.e., hourly or sub-hourly time scales. Specifically, these technologies have advanced knowledge of diel cycles produced by stream metabolism, concentration dynamics during storm events and assessment of uncertainty in load estimations (Aubert et al., 2014, Ferrant et al., 2013, Halliday et al., 2013, Pellerin et al., 2009, Rode et al., 2016, Schwientek et al., 2013, Shrestha et al., 2007, Wade et al., 2012). However, combined analysis of long-term and high-frequency water quality time series is rare, because high-frequency sensors are new technologies and are usually not set up near the long-term monitoring stations operated by environmental authorities.
The general objective of the present study was to investigate nitrate-N concentration dynamics in a stream at time scales ranging from 15 min to three decades, with the hypothesis that combining high-frequency and long-term analysis provides a basis to disentangle the factors controlling stream nitrate-N concentrations. Specifically, this study aimed to assess the respective influence of agricultural management and weather and climate variability on observed nitrate-N dynamics to evaluate the effectiveness of mitigation measures aimed at reducing nitrate-N pollution. The selected catchment (Weida, ~ 100 km2, eastern Germany) is of particular interest for this purpose because of its management history (as influenced by the reunification of Germany in the early 1990's, an ambitious programme of measures to reduce nitrate-N pollution in the late 1990's, and development of bioenergy crops since the 2000's), and because both high-frequency and long-term monitoring have been performed at the same location.
Section snippets
Study area and monitoring programme
Weida is a 99.5 km2 catchment located in eastern Germany, in the federal state of Thuringia. The Weida catchment is a sub-basin of the Saale River, which is a major tributary of the Elbe River. The Weida River flows into the Zeulenroda and Weida reservoirs, which were used for drinking water production from 1956 to 2012.
The climate is temperate continental, with mean long-term (1984–2004) annual precipitation of 628 mm (March–May: 152 mm; June–August: 210 mm; September–November: 142 mm;
Storm-event dynamics
Based on the criteria defined to detect storms from 15 min resolution discharge data, 26 events were identified during the 2005–2006 hydrological year (Fig. 2). Among them, 23 had continuous nitrate-N measurement and were selected for further analysis.
Fig. 3 represents discharge and concentration dynamics for each individual event as a function of time, as well as discharge - concentration plots. Table 1 summarises the variables used to describe nitrate-N storm event dynamics. Nitrate dilution
Seasonal variability in baseflow and storm nitrate dynamics
Both the 2005–2006 hydrological year and the long-term record exhibited a seasonal cycle in nitrate-N concentrations, with maxima during winter high flows and minima during summer low flows. Many catchments worldwide exhibit a positive concentration-discharge on a seasonal basis (e.g., Aubert et al., 2013, Martin et al., 2004, Mellander et al., 2014, Rodriguez-Blanco et al., 2015), also called an accretion pattern (Musolff et al., 2015). Previous studies have shown that an accretion pattern of
Conclusion
In this study, river nitrate dynamics in a 100 km2 catchment in eastern Germany were investigated from sub-hourly to decadal time scales. Sub-hourly resolution data during storm-events were analysed to identify periods of high leaching risk in the catchment: autumn, when high nitrate concentrations in the soil coincide with high drainage potential, and spring, during the fertilisation time window. It was assumed that these periods of high leaching risk led not only to the occurrence of few
Acknowledgements
The authors would like to thank the Thuringian Reservoir Administration for providing the data. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors
References (68)
- et al.
Nitrate leaching in intensive agriculture in northern France: effect of farming practices, soils and crop rotations
Agric. Ecosyst. Environ.
(2005) - et al.
A long-term view of nutrient transfers through the Seine river continuum
Sci. Total Environ.
(2007) - et al.
Long term change of nutrient concentrations of rivers discharging in European seas
Sci. Total Environ.
(2011) - et al.
Seasonal variability of NO3- mobilization during flood events in a Mediterranean catchment: the influence of intensive agricultural irrigation
Agric. Ecosyst. Environ.
(2015) - et al.
Integrating local knowledge and biophysical modeling to assess nitrate losses from cropping systems in drinking water protection areas
Environ. Model Softw.
(2015) - et al.
Assessing N emissions in surface water at the national level: comparison of country-wide vs. regionalized models
Sci. Total Environ.
(2013) - et al.
Using long time series of agricultural-derived nitrates for estimating catchment transit times
J. Hydrol.
(2015) - et al.
Soil-profile distribution of inorganic N during 6 years of integrated crop-livestock management
Soil Tillage Res.
(2013) - et al.
The role of climate on inter-annual variation in stream nitrate fluxes and concentrations
Sci. Total Environ.
(2010) - et al.
An analysis of long-term trends, seasonality and short-term dynamics in water quality data from Plynlimon, Wales
Sci. Total Environ.
(2012)